This project studies physiological and cellular aspects of neuronal calcium signaling, with long-range emphasis on postsynaptic responses in large central nervous system neurons. Neurons respond to synaptic stimuli with a rise in cytosolic free Ca2+ concentration ([Ca2+]i) that is strongly modulated by the activity of intracellular Ca stores. The latter activity plays an important role in spatio-temporally shaping Ca2+ signals that regulate important processes such as gene expression and LTP induction. We had earlier shown that stimulus-induced increases in [Ca2+]i in a variety of neurons induce large, reversible elevations in the concentration of calcium within mitochondria, which in turn has several important physiological and pathophysiological effects. We have now further explored the consequences of mitochondrial Ca2+ uptake, showing that in hippocampal pyramidal neurons this activity profoundly affects regulatory kinase cascades and programmed cell death. In hippocampal neurons large increases in [Ca2+]i activate several important kinases, whereas lower [Ca2+]i enhances protein phosphatase activity. This Ca2+-dependent rebalancing of the phosphorylation status of certain key enzymes, e.g., Ca/calmodulin-dependent kinases (CaMKs), controls the activity of pathways central to neuronal plasticity. We previously found that mitochondrial Ca accumulation mediated by strong Ca2+ entry leads to an increase in the production of superoxide radicals (O2-), and that this activity up-regulates the phosphorylation of CaMKII and CaMKIV-dependent CREB by inhibiting an array of serine/threonine protein phosphatases (PP1, PP2A and/or PP2B). Similarly, both mitochondrial and NADPH oxidase-derived O2- enhance Ca2+-dependent activation of the prototypical signaling kinase ERK, but only mitochondrial O2- operates by an analogous mechanism, i.e., inhibition of S/T protein phosphatases. Since O2- can also block protein tyrosine phosphatases (PTPs), recent investigations have focused on Raf-1, an up-stream mediator of ERK. A 100Hz/18s stimulus activated Raf-1, resulting in an increase in ERK. Inhibition of the Ca2+-dependent, non-receptor tyrosine kinases Pyk-2 and Src suppressed Raf-1 and ERK, indicating that in hippocampal neurons Raf-1 is up-regulated by Pyk-2 and Src. Inhibition of either mitochondrial or NADPH oxidase-derived O2- attenuated stimulus-induced phosphorylation of both Pyk-2 and Src, implying a mechanism involving O2- inhibition of PTPs. Up-regulation of Pyk-2 and Src phosphorylation by the PTP blockers vanadate and dephostatin, which are normally without effect because PTPs are already constitutively blocked by O2-, was observed when O2- from either source was inhibited. These results indicate that both mitochondrial and NADPH oxidase-derived O2- promote the Raf-1 pathway for ERK activation by inhibiting PTPs, and provide yet another example of physiological regulation by reactive oxygen species. We have continued to investigate mechanisms of mitochondrial involvement in excitotoxic death of cultured hippocampal neurons. We previously showed that delayed apoptotic death of neurons following NMDA overstimulation is mediated by mitochondrial calcium overload, which is variable from one mitochondria to another and correlates with mitochondrial injury. There is parallel variability among mitochondria with regard to swelling, membrane rupture, and loss of membrane potential, leading to the hypothesis that a proapoptotic signal is released only by a subpopulation of damaged mitochondria. Evidence that mitochondrial swelling was caused by Ca overload include: 1) a correlation between swelling and the level of elevated mitochondrial Ca; 2) the absence of mitochondrial swelling in zero Ca medium; and 3) the absence of mitochondrial swelling after substitution of extracellular Ca by Ba. Using EM immunocytochemistry we found that the known apoptogen cytochrome c, normally localized exclusively to mitochondria, was released into the cytoplasm following overstimulation but, importantly, also retained in many mitochondria. Even at 6h post-stimulation, clearly apoptotic neurons still retained cytochrome c in their mitochondria (which were, however, fewer in number). FCCP pretreatment, which decreased mitochondrial Ca accumulation and improved survival, prevented cytochrome c release but not mitochondrial swelling, indicating that both elevated mitochondrial Ca and swelling are essential for cytochrome c release. Our results indicate that neuronal apoptosis induced by excitotoxic stimulation involves release of apoptogens from only a subpopulation of mitochondria that have been injured by Ca overload. Consequently, undamaged mitochondria can continue to function normally and produce enough ATP to support downstream apoptotic reactions.